Image feature point matching method and device, electronic equipment and storage medium

Abstract

The invention discloses an image feature point matching method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining the position relation between a first image and a second image, and obtaining the depth estimation information of a target feature point in the first image; wherein the first image and the second image are acquired from a first visual angle and a second visual angle of the target scene by the image acquisition device, the first visual angle and the second visual angle are at least partially overlapped, and the position relationship represents relative poses when the image acquisition device acquires the image at the first visual angle and the second visual angle; determining an estimated matching point corresponding to the target feature point in the second image in combination with the position relationship and the depth information of the target feature point; and searching a target matching point matched with the target feature point in the second image based on the estimated matching point. In this way, the efficiency and precision of feature point matching can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer vision technology, and in particular to a method, device, electronic device and storage medium for matching image feature points. Background Art

[0002] The matching of image feature points is to extract feature points from multiple frames and match the feature points of different frames, in order to obtain the relative acquisition posture relationship between different image frames. At present, some mainstream matching methods mainly use posture priors to constrain the feature points to achieve feature point matching, but the polar constraint method requires searching for matching feature points within a larger image range, and the matching efficiency needs to be improved. Summary of the invention

[0003] The present application at least provides a method, device, electronic device and storage medium for matching image feature points.

[0004] In a first aspect, the present application provides a method for matching image feature points, the method comprising: obtaining a positional relationship between a first image and a second image, and obtaining depth information of a target feature point in the first image; wherein the first image and the second image are acquired by an image acquisition device from a first perspective and a second perspective of a target scene, respectively, and the first perspective and the second perspective at least partially overlap, and the positional relationship represents a relative posture when the image acquisition device acquires images at the first perspective and when the image is acquired at the second perspective; combining the positional relationship and the depth estimation information of the target feature point, determining an estimated matching point in the second image corresponding to the target feature point; based on the estimated matching point, searching for a target matching point matching the target feature point in the second image.

[0005] The method obtains depth estimation information of target feature points, combines the positional relationship between the first image and the second image and the depth information of the target feature points in the first image, and determines the estimated matching points in the second image corresponding to the target feature points, that is, determines the approximate positions of the matching points in the second image that match the target feature points in the first image. Therefore, the target matching points that match the target feature points are searched in the second image based on the estimated matching points, that is, the search for target matching points that match the target feature points is performed based on the approximate positions of the matching points that match the target feature points in the second image as a benchmark, so as to realize the search for target matching points in the second image with a benchmark, thereby improving the efficiency and accuracy of the matching point search, that is, improving the efficiency and accuracy of feature point matching, and reducing the probability of mismatching.

[0006] In some embodiments, the positional relationship and depth estimation information of the target feature point are combined to determine the estimated matching point in the second image corresponding to the target feature point, including: using the positional relationship and depth estimation information of the target feature point to back-project the target feature point into three-dimensional space to obtain the three-dimensional projection point corresponding to the target feature point; projecting the three-dimensional projection point to the second image to obtain the estimated matching point in the second image corresponding to the target feature point.

[0007] In some embodiments, searching the second image for target matching points that match the target feature points based on the estimated matching points includes: determining a search area in the second image based on the estimated matching points; and searching the search area for target matching points that match the target feature points.

[0008] In some embodiments, determining a search area in the second image based on the estimated matching points includes: acquiring a preset area including the estimated matching points, and using the preset area as the search area.

[0009] In some embodiments, obtaining a preset area including the estimated matching point includes: taking an area with the estimated matching point as the center and a preset length as the radius as the preset area.

[0010] In some embodiments, the image acquisition device is mounted on a drone, and the first perspective and the second perspective are overhead perspectives; obtaining depth estimation information of the target feature point in the first image includes: obtaining the flight altitude information of the drone when the image acquisition device takes the first image, and using the flight altitude information as the depth estimation information of the target feature point.

[0011] According to a second aspect of the present application, there is provided a device for matching image feature points, the device comprising an acquisition module, a determination module and a search module; the acquisition module is used to acquire the projection relationship between a first image and a second image, and acquire depth estimation information of a target feature point in the first image; wherein the first image and the second image are acquired by an image acquisition device in the air from a first perspective and a second perspective of a target scene, respectively, and the first perspective and the second perspective at least partially overlap, and the positional relationship represents the relative posture of the image acquisition device when acquiring the image at the first perspective and when acquiring the image at the second perspective; the determination module is used to determine an estimated matching point corresponding to the target feature point in the second image in combination with the projection relationship and the depth estimation information of the target feature point; the search module is used to search for a target matching point matching the target feature point in the second image based on the estimated matching point.

[0012] A third aspect of the present application provides an electronic device, which includes a memory and a processor, wherein the memory stores program instructions, and the processor is used to execute the program instructions to implement the above-mentioned image feature point matching method.

[0013] A fourth aspect of the present application provides a computer-readable storage medium, which is used to store program instructions, and the program instructions can be executed to implement the above-mentioned image feature point matching method. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a flow chart of an embodiment of a method for matching image feature points provided by the present application;

[0015] Figure 2 is a schematic diagram of an embodiment of a first image and a second image provided by the present application;

[0016] Figure 3 is a schematic diagram of a flow chart of an embodiment of obtaining a positional relationship between a first image and a second image provided by the present application;

[0017] Figure 4 yes Figure 1 The flowchart of step S12 is shown as an embodiment;

[0018] Figure 5 yes Figure 1 The flowchart of step S13 is shown as an embodiment;

[0019] Figure 6 It is a structural schematic diagram of an embodiment of a device for matching image feature points provided by the present application;

[0020] Figure 7 It is a structural schematic diagram of an embodiment of an electronic device provided by the present application;

[0021] Figure 8 It is a structural schematic diagram of an embodiment of a computer-readable storage medium provided by the present application. DETAILED DESCRIPTION

[0022] The scheme of the embodiment of the present application is described in detail below in conjunction with the drawings of the specification.

[0023] In the following description, for the purpose of explanation rather than limitation, specific details such as specific system structures, interfaces, and technologies are provided to facilitate a thorough understanding of the present application.

[0024] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the objects associated before and after are in an "or" relationship. In addition, "many" in this article means two or more than two. In addition, the term "at least one" in this article means any combination of at least two of any one or more of a plurality of, for example, including at least one of A, B, and C, can mean including any one or more elements selected from the set consisting of A, B, and C.

[0025] See also Figure 1 , Figure 1 FIG. 1 is a flow chart of an embodiment of the method for matching image feature points provided by the present application. It should be noted that if there are substantially the same results, the embodiments of the present application are not limited to the above embodiments. Figure 1 The process sequence shown is limited. Figure 1 As shown, this embodiment includes:

[0026] Step S11: obtaining a positional relationship between the first image and the second image, and obtaining depth estimation information of target feature points in the first image.

[0027] The method of this embodiment is used to match feature points between multiple frames of images, so that the relative relationship between the acquisition devices when acquiring different image frames can be determined based on the matched feature points. It should be noted that the matching of feature points is to find the same feature of the same object in different images, which is an important method for finding information associations between different images in the image field; because each feature point has a descriptor that uniquely identifies its identity and characteristics, feature point matching is actually to find two feature points with similar descriptors in two images.

[0028] In this embodiment, the positional relationship between the first image and the second image is obtained, and depth estimation information of the target feature points in the first image is obtained; wherein the first image and the second image are acquired by an image acquisition device from a first perspective and a second perspective of the target scene, respectively, and the first perspective and the second perspective at least partially overlap, and the positional relationship between the first image and the second image represents the relative posture of the image acquisition device when acquiring the image at the first perspective and when acquiring the image at the second perspective.

[0029] Among them, the specific angles of the first perspective and the second perspective are not limited, and can be specifically set according to actual use needs; it should be noted that the first perspective and the second perspective at least partially overlap, indicating that at least part of the image content of the first image captured at the first perspective and the second image captured at the second perspective is the same. In addition, the target scenes described in this article include but are not limited to public places (such as parks, squares, office places, etc.), residential areas (such as bedrooms, living rooms, kitchens, etc.), and tourist areas (such as forests, beaches, amusement parks, etc.), which are not specifically limited here.

[0030] In one embodiment, the first image and the second image are acquired by an image acquisition device in the air from a first perspective and a second perspective of a target scene, respectively, wherein the difference between the posture of the image acquisition device when acquiring the first image at the first perspective and the posture when acquiring the second image at the second perspective is the positional relationship between the first image and the second image.

[0031] In a specific embodiment, the image acquisition device is mounted on a drone to realize image acquisition of the target scene in the air. That is to say, the feature point matching method provided by this application is specifically used in the field of drones to assist subsequent drones in navigation, AR effect drawing, etc. Of course, in other embodiments, the first image and the second image can also be obtained by the image acquisition device on the horizontal plane from the first perspective and the second perspective of the target scene, respectively, which is not specifically limited here.

[0032] In one embodiment, the image acquisition device may be a depth camera, and the depth camera is used to estimate the depth information of the feature points in the acquired image, so that the depth information of the target feature points in the first image can be obtained as the depth estimation information. The specific type of the depth camera is not limited, and can be specifically set according to actual use needs, such as the depth camera is a structured light depth camera, a time of flight (TOF) depth camera, etc.

[0033] In other embodiments, the image acquisition device may also be a laser scanner, which uses the principle of laser ranging to record the three-dimensional coordinates, reflectivity, texture and other information of a large number of dense points on the surface of the measured object, and can restore and build a three-dimensional model of the measured object and various data such as lines, surfaces, and volumes. Currently, in other embodiments, the image acquisition device may also be an ordinary camera, a binocular camera, etc., which is not specifically limited here.

[0034] In a specific embodiment, the image acquisition device is mounted on a drone, and the first and second perspectives are overhead perspectives. At this time, the depth estimation information of the target feature points in the first image is obtained, specifically: the flight altitude information of the drone when the image acquisition device takes the first image is obtained, and the flight altitude information is used as the depth estimation information of the target feature points. Since the image acquisition device is mounted on a drone, during the flight of the drone, the drone takes a bird's-eye view of the ground, and the height of other objects on the ground, such as houses, is almost negligible for the drone; therefore, when the image acquisition device is mounted on a drone and the drone takes a bird's-eye view of the ground, the flight altitude information of the drone can be directly used as a rough estimate of the depth information of each point on the image frame, thereby obtaining the depth estimation information of the target feature points in the first image.

[0035] Step S12: Determine an estimated matching point in the second image corresponding to the target feature point by combining the positional relationship and the depth estimation information of the target feature point.

[0036] In this embodiment, the estimated matching point corresponding to the target feature point in the second image is determined by combining the positional relationship and the depth estimation information of the target feature point. That is, by combining the positional relationship between the first image and the second image and the depth estimation information of the target feature point in the first image, the approximate position of the matching point in the second image that matches the target feature point in the first image can be determined.

[0037] For example, if Figure 2 As shown, Figure 2 is a schematic diagram of an embodiment of the first image and the second image provided by the present application; the target feature point in the first image is Figure 2 The point p1 shown, combined with the positional relationship between the first image and the second image and the depth estimation information of point p1, determines that the estimated matching point corresponding to the target feature point in the second image is point p2. Wherein O1 represents the optical center position when the image acquisition device acquires the first image at the first perspective, and O2 represents the optical center position when the image acquisition device acquires the second image at the second perspective. According to the relative posture between the first image and the second image, the relative posture of the two optical center positions O1 and O2 can be determined. The "depth estimation" point represents the position of the three-dimensional space point corresponding to p1 estimated according to the depth estimation information. The "depth estimation point" is projected onto the second image plane in combination with the relative posture of O1 and O2 to obtain the estimated matching point p2.

[0038] Step S13: Based on the estimated matching points, searching the second image for target matching points that match the target feature points.

[0039] In this embodiment, the target matching point that matches the target matching point is searched in the second image based on the estimated matching point. Since the estimated matching point is the approximate position of the matching point in the second image that matches the target feature point in the first image, the target matching point that matches the target feature point is searched in the second image based on the estimated matching point, that is, the target matching point that matches the target feature point is searched based on the approximate position of the matching point in the second image that matches the target feature point as a reference, so as to realize the search of the target matching point in the second image in a directional and referenced manner, thereby improving the efficiency and accuracy of feature point matching and reducing the probability of mismatching.

[0040] In one embodiment, a target matching point matching the target feature point is searched in a search area determined based on the estimated matching point. It is understandable that in other embodiments, a target matching point matching the target feature point may also be searched on a search line determined based on the estimated matching point, which is not specifically limited here.

[0041] In the above-mentioned implementation, the estimated matching point corresponding to the target feature point in the second image is determined in combination with the positional relationship between the first image and the second image and the depth estimation information of the target feature point in the first image, and the target matching point matching the target feature point is searched in the second image based on the estimated matching point. Therefore, since the estimated matching point is the approximate position of the matching point in the second image that matches the target feature point in the first image, searching for the target matching point matching the target feature point in the second image based on the estimated matching point is to search for the target matching point matching the target feature point based on the approximate position of the matching point matching the target feature point in the second image as a benchmark, so as to realize the search for the target matching point in the second image with a benchmark, thereby improving the efficiency and accuracy of the matching point search, namely improving the efficiency and accuracy of the feature point matching, and reducing the probability of mismatching. At the same time, since the depth estimation information of the target feature point can be a rough estimate, there is no need to perform high-precision depth detection, which reduces the difficulty of matching the target feature point.

[0042] When applied to a scene where a drone is photographing the ground, since the ground images taken by the drone contain more repeated textures, large-scale repeated texture matching may cause large errors. The method of this embodiment narrows the search range of the target matching point, thereby reducing the probability of such errors, thereby improving the matching accuracy of the feature points.

[0043] See also Figure 3 , Figure 3 1 is a flow chart of an embodiment of obtaining the positional relationship between the first image and the second image provided by the present application. It should be noted that if there are substantially the same results, the embodiments of the present application are not limited to Figure 3 The process sequence shown is limited. Figure 3As shown, in this embodiment, the process of obtaining the positional relationship between the first image and the second image includes:

[0044] Step S31: acquiring a first posture of the image acquisition device when shooting at a first viewing angle, and acquiring a second posture of the image acquisition device when shooting at a second viewing angle.

[0045] Since the first image and the second image are respectively captured by the same image acquisition device at the first viewing angle and the second viewing angle, the positional relationship (conversion relationship) between the first image and the second image is actually the change relationship of the shooting posture of the image acquisition device. Therefore, in this embodiment, the first posture of the image acquisition device when shooting at the first viewing angle is first obtained, and the second posture of the image acquisition device when shooting at the second viewing angle is obtained.

[0046] Step S32: Determine a conversion relationship between the first posture and the second posture as a positional relationship.

[0047] In this embodiment, the conversion relationship between the first posture and the second posture is determined as the positional relationship between the first image and the second image. Since the first image and the second image are respectively captured by the same image acquisition device at the first viewing angle and the second viewing angle, the positional relationship (conversion relationship) between the first image and the second image is actually the posture conversion relationship of different shooting postures of the image acquisition device.

[0048] See also Figure 4 , Figure 4 yes Figure 1 It should be noted that if there are substantially the same results, the embodiments of the present application are not limited to the above embodiments. Figure 4 The process sequence shown is limited. Figure 4 As shown, the process of step S12 in this embodiment includes:

[0049] Step S41: using the positional relationship and depth estimation information of the target feature point, the target feature point is back-projected into the three-dimensional space to obtain the three-dimensional projection point corresponding to the target feature point.

[0050] In this embodiment, the target feature points are back-projected into three-dimensional space using the positional relationship between the first image and the second image and the depth estimation information of the target feature points in the first image to obtain the three-dimensional projection points corresponding to the target feature points. Specifically, the three-dimensional information of the target feature points in the first image can be obtained based on the depth estimation information of the target feature points in the first image; then, based on the conversion system between the first image and the second image, the target feature points in the first image are converted to obtain the three-dimensional projection points corresponding to the target feature points.

[0051] Step S42: Projecting the three-dimensional projection point onto the second image to obtain an estimated matching point in the second image corresponding to the target feature point.

[0052] In this embodiment, the three-dimensional projection point is projected onto the second image to obtain the estimated matching point corresponding to the target feature point in the second image. The three-dimensional projection point corresponding to the target feature point is projected onto the second image to obtain the estimated matching point corresponding to the target feature point in the second image.

[0053] See also Figure 5 , Figure 5 yes Figure 1 FIG. 1 is a flow chart of an embodiment of step S13. It should be noted that if there are substantially the same results, this embodiment does not necessarily use the same method. Figure 5 The process sequence shown is limited. Figure 5 As shown, in an embodiment of the present application, in a search area determined based on estimated matching points, a process of searching for a target matching point that matches a target feature point specifically includes:

[0054] Step S51: Determine a search area in the second image based on the estimated matching points.

[0055] In this embodiment, a search area is determined in the second image based on the estimated matching points. That is, based on the position in the second image that matches the target feature point, an area is determined in the second image to search for the target matching point that matches the target feature point; on the one hand, since the estimated matching point is the approximate position of the matching point in the second image that matches the target feature point in the first image, the search area determined in the second image based on the estimated matching point is likely to contain a target matching point that matches the target feature point, so that the target matching point that matches the target feature point can be searched more accurately in the subsequent process, thereby improving the feature point matching accuracy; on the other hand, the search area is smaller than the second image area, which significantly reduces the area for searching the target matching point, thereby improving the feature point matching efficiency while improving the feature point matching accuracy in the subsequent process.

[0056] The position of the estimated matching point is the approximate position of the matching point in the second image that matches the target feature point in the first image. It is highly likely that there are target matching points matching the target feature point at and around the estimated matching point. Therefore, in order to improve the accuracy and efficiency of subsequent searches for target matching points matching the target feature point, that is, in order to improve the efficiency and accuracy of feature point matching, in one embodiment, a preset area including the estimated matching point is obtained, and the preset area is used as the search area.

[0057] In one specific embodiment, the size of the preset area including the estimated matching point is fixed, and the specific size is not limited here, and can be set according to actual use needs. In other specific embodiments, the size of the preset area including the estimated matching point is flexibly changed; specifically, the matching degree between the estimated matching point and the target feature point is determined. When the matching degree between the estimated matching point and the target feature point is greater than the matching threshold, it indicates that the matching degree between the estimated matching point and the target feature point is high, and there is a high possibility that there are target matching points matching the target feature point in the vicinity of the estimated matching point. Therefore, a smaller preset area including the estimated matching point is obtained, and vice versa, a larger preset area including the estimated matching point is obtained.

[0058] In a specific embodiment, Figure 2 As shown, the area with the estimated matching point as the center and the preset length as the radius is used as the preset area; specifically, the area constructed with the estimated matching point p2 as the center and the preset length l2 as the radius is used as the search area determined in the second image. In other specific embodiments, the area with the estimated matching point as the center and the preset length as the side length can also be used as the preset area. In other specific embodiments, the area with the estimated matching point as the center, the first preset length as the length, and the second preset length as the width can also be used as the preset area, which is not specifically limited here.

[0059] Step S52: Searching for target matching points that match the target feature points in the search area.

[0060] In this embodiment, a target matching point that matches the target feature point is searched out from the search area. Based on the approximate position of the matching point that matches the target feature point in the second image, an area for searching the target matching point that matches the target feature point is determined in the second image, and the target matching point that matches the target feature point is searched; on the one hand, since the estimated matching point is the approximate position of the matching point in the second image that matches the target feature point in the first image, the search area determined in the second image based on the estimated matching point has a high possibility of a target matching point that matches the target feature point, so that the target matching point that matches the target feature point can be searched out more accurately, and the feature point matching accuracy is improved; on the other hand, the search area is smaller than the second image area, which significantly reduces the area for searching the target matching point, which can improve the feature point matching accuracy while improving the feature point matching efficiency.

[0061] See also Figure 6 , Figure 6It is a structural schematic diagram of an embodiment of a matching device for image feature points provided by the present application. The matching device 60 for image feature points includes an acquisition module 61, a determination module 62 and a search module 63. The acquisition module 61 is used to acquire the positional relationship between the first image and the second image, and to acquire the depth estimation information of the target feature points in the first image; wherein the first image and the second image are acquired by an image acquisition device from a first perspective and a second perspective of the target scene, respectively, and the first perspective and the second perspective at least partially overlap, and the positional relationship represents the relative posture when the image acquisition device acquires the image at the first perspective and when it acquires the image at the second perspective; the determination module 62 is used to determine the estimated matching points corresponding to the target feature points in the second image in combination with the positional relationship and the depth estimation information of the target feature points; the search module 63 is used to search for target matching points matching the target feature points in the second image based on the estimated matching points.

[0062] Optionally, the determination module 62 is used to combine the positional relationship and the depth estimation information of the target feature point to determine the estimated matching point corresponding to the target feature point in the second image in the following manner: using the positional relationship and the depth estimation information of the target feature point, the target feature point is back-projected into three-dimensional space to obtain the three-dimensional projection point corresponding to the target feature point; the three-dimensional projection point is projected onto the second image to obtain the estimated matching point corresponding to the target feature point in the second image.

[0063] Optionally, the search module 63 is used to search for target matching points that match the target feature points in the second image in the following manner: based on the estimated matching points, determine a search area in the second image; and search for target matching points that match the target feature points in the search area.

[0064] Optionally, the search module 63 is used to determine a search area in the second image in the following manner: obtain a preset area including the estimated matching points, and use the preset area as the search area.

[0065] Optionally, the search module 63 is used to obtain a preset area including the estimated matching point in the following manner: a region with the estimated matching point as the center and a preset length as the radius is taken as the preset area.

[0066] Optionally, the image acquisition device is mounted on a drone, and the first perspective and the second perspective are overhead perspectives; the acquisition module 61 is used to obtain depth estimation information of target feature points in the first image in the following manner: obtain the flight altitude information of the drone when the image acquisition device takes the first image, and use the flight altitude information as the depth estimation information of the target feature points.

[0067] See also Figure 7 , Figure 7: is a schematic diagram of the structure of an embodiment of an electronic device provided by the present application. The electronic device 70 includes a memory 71 and a processor 72 coupled to each other, and the processor 72 is used to execute program instructions stored in the memory 71 to implement the steps of any of the above-mentioned image feature point matching method embodiments. In a specific implementation scenario, the electronic device 70 may include but is not limited to: a microcomputer, a server, and in addition, the electronic device 70 may also include a mobile device such as a laptop computer and a tablet computer, which is not limited here.

[0068] Specifically, the processor 72 is used to control itself and the memory 71 to implement the steps of the above-mentioned matching method embodiment of any image feature point. The processor 72 can also be called a CPU (Central Processing Unit). The processor 72 may be an integrated circuit chip with signal processing capabilities. The processor 72 can also be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field-programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. In addition, the processor 72 can be implemented by an integrated circuit chip.

[0069] See also Figure 8 , Figure 8 It is a structural diagram of an embodiment of a computer-readable storage medium provided by the present application. The computer-readable storage medium 80 of the embodiment of the present application stores a program instruction 81, and when the program instruction 81 is executed, it implements the method provided by any embodiment of the image feature point matching method of the present application and any non-conflicting combination. Among them, the program instruction 81 can form a program file and be stored in the above-mentioned computer-readable storage medium 80 in the form of a software product, so that a computer device (which can be a personal computer, a server, or a network device, etc.) executes all or part of the steps of each implementation method of the present application. The aforementioned computer-readable storage medium 80 includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk, or a terminal device such as a computer, a server, a mobile phone, and a tablet.

[0070] If the technical solution of this application involves personal information, the product using the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using the technical solution of this application has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, the personal information processing rules are notified by obvious signs / information, and the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

[0071] The above description is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for matching image feature points, characterized in that: The method comprises: Acquire a positional relationship between a first image and a second image, and acquire depth estimation information of a target feature point in the first image; wherein the first image and the second image are acquired by an image acquisition device from a first perspective and a second perspective of a target scene, respectively, and the first perspective and the second perspective at least partially overlap, and the positional relationship represents a relative posture when the image acquisition device acquires an image from the first perspective and when the image is acquired from the second perspective; Determine an estimated matching point in the second image corresponding to the target feature point by combining the positional relationship and the depth estimation information of the target feature point; Based on the estimated matching points, a target matching point matching the target feature point is searched in the second image.

2. The method according to claim 1, characterized in that The determining, in combination with the positional relationship and the depth estimation information of the target feature point, an estimated matching point corresponding to the target feature point in the second image includes: Using the positional relationship and the depth estimation information of the target feature point, the target feature point is back-projected into a three-dimensional space to obtain a three-dimensional projection point corresponding to the target feature point; The three-dimensional projection point is projected onto the second image to obtain an estimated matching point in the second image corresponding to the target feature point.

3. The method according to claim 1, characterized in that The step of searching the second image for a target matching point that matches the target feature point based on the estimated matching point includes: Based on the estimated matching points, determining a search area in the second image; A target matching point matching the target feature point is searched from the search area.

4. The method according to claim 3, characterized in that The step of determining a search area in the second image based on the estimated matching points includes: A preset area including the estimated matching point is obtained, and the preset area is used as the search area.

5. The method according to claim 4, characterized in that The obtaining of the preset area including the estimated matching point comprises: An area with the estimated matching point as the center and the preset length as the radius is used as the preset area.

6. The method according to claim 1, characterized in that The image acquisition device is mounted on a drone; the first viewing angle and the second viewing angle are overhead viewing angles, and the depth estimation information of the target feature points in the first image is obtained, including: The flying height information of the UAV when the image acquisition device takes the first image is obtained, and the flying height information is used as the depth estimation information of the target feature point.

7. A device for matching image feature points, characterized in that: The device comprises: An acquisition module, configured to acquire a positional relationship between a first image and a second image, and acquire depth estimation information of a target feature point in the first image; wherein the first image and the second image are acquired by an image acquisition device from a first perspective and a second perspective of a target scene, respectively, and the first perspective and the second perspective at least partially overlap, and the positional relationship represents a relative posture when the image acquisition device acquires an image at the first perspective and when the image is acquired at the second perspective; A determination module, configured to determine an estimated matching point in the second image corresponding to the target feature point by combining the positional relationship and the depth estimation information of the target feature point; A search module is used to search for a target matching point matching the target feature point in the second image based on the estimated matching point.

8. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores program instructions, and the processor is used to execute the program instructions to implement the method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program instructions, and the program instructions can be executed to implement the method according to any one of claims 1 to 6.

Images